UNIVERSALISM AND THE PROBLEM OF HELL. reason to believe in it, revelation must provide it. Yet, McTaggart continued, the infliction of

Transcription

1 UNIVERSALISM AND THE PROBLEM OF HELL A very concise formulation of the problem of hell is due to the British metaphysician 1 John McTaggart. He argued that if there is a hell, one could have no good reason to believe it. He reasoned that there is no good empirical evidence to believe in hell, so that if there is a good reason to believe in it, revelation must provide it. Yet, McTaggart continued, the infliction of hell is very wicked, and so anyone who would send one to hell must be vile indeed. In such a case, one could have no good reason to trust such an individual concerning anything of importance to our well-being, for as anyone familiar with the games played by schoolyard bullies knows, there is no telling why a vile person would say that something is good or bad for one. It may be for amusement, it may be to see one suffer, or it may be for any of a host of other reasons which are compatible with the falsity of what is being said. So if anyone, including God, tells us that there is a hell to shun, McTaggart claims that we could have no good reason to believe that such an individual is telling us the truth. What such a person says, to put the point succinctly, undercuts the reliability of the testimony. We can put this argument in the form of a dilemma which, in honor of its source, I will call "McTaggart's Dilemma". Either there is good reason to believe in hell or there is not. If there is no good reason for belief in hell, no one should believe in it. After all, without some reason to believe in hell, the option of believing in hell is no better than any other merely conceivable catastrophe, (such as the catastrophe of accepting the view that committing suicide immediately is the only way to escape an eternal medieval torture chamber). The point is that there are too many mere conceivables that our lives end in ruin and too many merely conceivable routes to avoid these conceivables to make belief in any one of them plausible at all. So, if the doctrine of hell is to be respectable, it must be because there is good reason to believe in it. The difficulty for this option, according to McTaggart, is that the only kinds of reasons 60

2 there might be are self-defeating. For the only evidence there might be for hell is on the basis of revelation from the one who can consign one to hell, and yet no one of that sort who claims that hell exists and outlines conditions for avoiding hell can be trusted. So, on either horn of the dilemma, the conclusion to be drawn, according to McTaggart, is that belief in hell is not intellectually respectable. McTaggart's Dilemma shows, I will assume here, that an adequate version of Christianity must include a doctrine of hell on which it would not be wrong for someone to end up in hell. That is, I will assume here that there is no adequate version of Christianity without some doctrine of hell. That may seem to be a large assumption, but since I have argued for it elsewhere (and argued that it is, in fact, not a large assumption when hell is understood in a 2 minimal way in terms of the logical contrast to heaven), I ll merely assume it here. For those inclined against this assumption, the argument here can be conceived in terms of answering the following conditional question: if Christianity is to contain a doctrine of hell, what must that doctrine be like? The argument McTaggart gives would not be rhetorically interesting were it not for the fact that the traditional conception of hell is plausibly held to succumb to that dilemma. According to traditional Christian teaching, hell is a place where some people are punished eternally with no possibility of escape. We can analyze this traditional conception, which I term "the strong view," into four separate components: (H1) The Anti-Universalism Thesis: some persons are consigned to hell; (H2) The Existence Thesis: hell is a place where people exist, if they are consigned there; (H3) The No Escape Thesis: there is no possibility of leaving hell, there is nothing one can do, and change, or become in order to get out of hell, once one is consigned there; 61

3 (H4) The Retribution Thesis: the justification for and purpose of hell is retributive in nature, hell being constituted so as to mete out punishment to those whose earthly lives and behavior warrant it. As already noted, this doctrine has appeared morally objectionable to many individuals, and I do not intend to dispute that claim here. My interest is rather in an alternative which has become increasingly popular over the past few centuries. This position, universalism, maintains that the basic structure of hell is as the strong view maintains, but claims that the population statistics the strong view implies are mistaken. Where the strong view goes wrong, universalists maintain, is in thinking that hell is populated at all. All will in the end be saved, according to universalism, and once we accept their viewpoint, universalists maintain, the problem of hell disappears. Universalist views, though traceable to the Alexandrian school of theologians founded by Clement and Origen, have become increasingly popular in the last three to four hundred years. 3 Though orthodoxy has never wavered since condemning universalism at the Fifth General 4 Council at Constantinople in 553 A.D., a revival of interest in universalism began in the late seventeenth century with the Cambridge Platonists and carried into the eighteenth and nineteenth centuries in the work of Swedenborg, Tennyson, Kant, Schleiermacher, Ritschl, and in the rise of 5 the Unitarian and Universalist denominations in America. Universalism continues into the twentieth century, though the rise of the influence of existentialism clouds what might otherwise 6 have been a complete embracing by many of the most influential theologians of our century. In addition, the demythologizing programme of Bultmann and the reinterpretation of salvation in terms relating to social, political, and economic affairs have all done their part both to undermine the traditional understanding of hell and to mask universalist tendencies. In those for whom the afterlife is taken seriously and one s fate in it of importance, however, it is fair to say that universalism presents by far the most attractive alternative to the strong view. 62

4 Just endorsing the view, however, isn t sufficient in itself to provide a solution to McTaggart s Dilemma. The view must be developed in a way that implies some mistake in this dilemma but which is also satisfactory philosophically and theologically. The prospects for universalism are intertwined with such thorny problems as the nature of human freedom and the conditions for salvation, and no breezy and shallow sentiment in favor of the idea that everything will work out for the best in the end can be thought adequate here. We can begin entry into these issues by first distinguishing versions of universalism in terms of modal status. Contingent universalism holds that, though it is possible that a person end up in hell as described by the conjunction of (H2)-(H4), as a matter of contingent fact every human being will end up in heaven. Necessary universalism holds that it is not only true but necessarily true that every human being will end up in heaven; it is simply impossible that anyone be damned. I will argue that both versions fail to solve the problem of hell; that contingent universalism is too weak to solve the problem and necessary universalism is too strong. I will then turn to the question of whether there is any adequate middle ground between these two positions. 1. Contingent Universalism According to contingent universalism, the possibility exists that some people end up in hell, but as a matter of contingent fact, no one will. In the end, God's saving grace and power win out over the forces of evil, and the entire created order is reconciled to God through Christ Jesus. This position is quite common among contemporary theologians. One example is John Macquarrie. His understanding of hell is in terms of annihilation; he says, "If heaven is fullness of being and the upper limit of human existence, hell may be taken as loss of being and the lower 63

5 limit... This utter limit of hell would be annihilation, or at least the annihilation of the 7 possibility for personal being." Given this understanding of hell, Macquarrie goes on to suggest a version of universalism. He says, "If God is indeed absolute letting-be, and if his letting-be has power to overcome the risks of dissolution, then perhaps in the end... no individual existence 8 that has been called out of nothing will utterly return to nothing..." To drive home his rejection of traditional conceptions of hell, he says, "Needless to say, we utterly reject the idea of a hell where God everlastingly punishes the wicked, without hope of deliverance. Even earthly penologists are more enlightened nowadays." 9 So Macquarrie holds that the strong view of hell and its associates are at best unenlightened (elsewhere he terms such views "barbarous"), that hell is best thought of in terms of annihilation, though the concept is a limiting one, in the realm of mere possibility rather than the destiny of some human beings. Of special interest in the present context is the juxtaposition of universalism with the strong view of hell and its associates. This juxtaposition suggests the possibility that the inadequacies found in traditional conceptions of hell can be thought of as being overcome by a doctrine of universalism. It is the motivation behind the affirmation of universalism that is important here, a motivation having to do with the need to solve the moral problem of hell. This theoretical motivation can be found even in some of the more conservative Christian groups. Clark Pinnock, a conservative evangelical theologian, claims, If the doctrine of hell is taken to mean (as it so often is) that God raises up the wicked to everlasting existence for the express purpose of inflicting upon them endless pain and torment, universalism will become practically irresistible in its appeal to sensitive Christians... If the only options are torment and unversalism, then I would expect large 0 numbers of sensitive Christians to choose universalism. 1 64

6 Pinnock goes on to defend the annihilation view, but in the present context, what is of greater significance is that he sees in universalism a solution to the moral problems facing the strong view of hell and its associates. According to Pinnock, if the conservative Christian faces the options of a morally problematic view of hell on the one hand and universalism on the other, it would be fully understandable if the conservative Christian opted to embrace universalism. The pressing question in the context of McTaggart s Dilemma is how endorsing contingent universalism is supposed to block the dilemma. We may presume, I think, that the idea is as follows. Central to McTaggart s Dilemma is the idea that sending someone to hell would be an incredibly wicked thing to do, and so if no one is sent to hell, the wickedness claim that is central to that dilemma should be rejected. So the dilemma disappears. The problem with contingent universalism is that the wickedness claim involved in McTaggart s Dilemma doesn t turn on population issues concerning hell. The claim in the dilemma is that the infliction of hell involves wickedness. What follows from such a claim is that if God sent someone to hell, he would have done a wicked thing. But suppose God doesn t actually send anyone to hell, as contingent universalists claim. Even so, the wickedness charge doesn t disappear. For God is not only perfectly and wholly good in this world, He would be perfectly and wholly good no matter what the course of affairs took. He is not just contingently morally perfect; rather, His moral perfection is essential to His character. So, suppose no one actually is sent to hell, but had things been slightly different, God would have sent someone to hell. Then the dilemma becomes pressing again, for such a counterfactual possibility would involve wickedness, according to McTaggart, and nothing the contingent universalist has claimed has any power to show that McTaggart is mistaken. If the wickedness claim remains unscathed by contingent universalism, the trust issue about whether to trust the word of someone who tells us about hell is still in place. Since we are assuming that a doctrine of hell is central to Christianity, contingent universalism would leave us with McTaggart s Dilemma in place and 65

7 thus with no good reason to think that an essential part of Christianity is true. So, if the worry is that the infliction of hell is wicked, affirming contingent universalism will not solve that problem. At best, it only modally masks it, with the hope that McTaggart doesn t have the insight to see that it is only a mask. 2. Necessary Universalism If contingent universalism fails to engage the dilemma by undermining the claim that the infliction of hell involves wickedness, necessary universalism appears much better off. Necessary universalism holds that it is impossible that hell is inflicted on anyone, and if that is true, no threat to God s perfect goodness remains. One might try to reinstate McTaggart s Dilemma by insisting that it nonetheless remains true that if God were to send someone to hell, a wicked thing would have been done, and that if this conditional is true, there is still a worry about anyone warning us to avoid something that would be wicked to impose on us. In support of this claim, we might be reminded that threatening to do something wicked is itself wicked. Here we need to be reminded of the defeasible relationships that typically hold in moral and epistemic reasoning. To learn that a given threat is a threat to do something wicked is a defeasible reason for thinking that the threat itself is wicked, but it is no logical guarantee that the threat itself is wicked. Adopting a policy of mutually assured nuclear destruction is to threaten to do something wicked, but if it is the only way to prevent a nuclear attack by an enemy, the consequences of such an attack are so severe that only the most recalcitrant deontologists will insist that it is not justified. If only rabid Kantians can press the point, necessary universalists are on safe enough ground that we can push forward considering their proposal, leaving what to say to, and what to do with, rabid Kantians to another time and place (perhaps hoping for enough insightful critique from gifted ethicists to relieve us of further 66

8 argumentative burden). Further, we might also note that the worry about the wickedness of hell is clearly a worry deriving from the consequences of suffering eternal damnation rather than some feature more plausibly appealed to by strict Kantianism, so relying on such a strong deontologism at this point will undercut the basis of the dilemma (to say nothing of the fact that cases in which real disaster looms unless Kantian strictures are violated look like clear counterexamples to those strictures). So necessary universalism, unlike contingent universalism, offers hope of avoiding McTaggart s Dilemma. Among the philosophically unsophisticated, one often hears the worry that such a view compromises the very heart of Christianity, according to which we are all headed for hell apart from the redemptive work of God in Christ. But such a worry is confused. The claim in question that we are all headed for hell apart from God s intervention is no less 11 true should the intervention in question be necessitated by God s nature. So we need not balk at necessary universalism on these grounds. There are grounds, however, for worrying that the position of necessary universalism is indefensible. Some worries are theological in nature, claiming that an adequate soteriology requires a role for faith that the appropriate kind of faith can obtain only if one has knowledge of the existence of a first century Jewish man named Jesus. There is even a proof-text for the view: "Salvation is found in no one else, for there is no other name under heaven given to me by which 12 we must be saved." According to such individuals, it would not be sufficient, for example, to have faith in God himself in order to be saved, nor would it be sufficient to recognize one's own inability to achieve salvation and to trust in what God has and will do for one. The conceptual contents of faith in these latter examples are too general; in order to be saving faith, the conceptual content of faith must involve particular reference to a historical individual of the first century, Jesus of Nazareth. 67

9 Once this soteriological position is adopted, necessary universalism appears very hard to defend. For it is hard to imagine how any such piece of human knowledge could be necessary. In fact, it appears not even to be true, say nothing of necessarily true, that every human being knows who Jesus is. There are, however, serious difficulties with this soteriological position and the use of this passage as a proof-text for the view. For one thing, this objection must assume that the knowledge in question has to be knowledge in this life, prior to death. Without this assumption, it is much harder to insist that such knowledge would have to be contingent. For another thing, those who hold this view would also agree with the apostle Paul that Abraham was righteous by faith, and yet it is quite implausible if not nonsensical to hold that Abraham was aware of the existence of a first century Jewish man named Jesus. A full exploration of this topic is required if our goal is a complete soteriology; but given our present purposes, it is sufficient to note that the view of faith that requires conceptual content referring to Jesus of Nazareth is far from obviously correct (even given quite conservative assumptions about doctrinal matters). Exactly what kind of conceptual content is required is an interesting question to explore, but it is beyond 13 our present purposes. Even if there is some way to formulate this objection to get around the problems it faces, I will not here rest a rejection of this universalist solution to McTaggart s Dilemma on this sort of objection, for I believe there is a much more serious one. This more serious problem arises from what I will call "the free will argument." In succinct form, it claims that God cannot guarantee the presence of all in heaven without being willing to violate the freedom of some individuals to choose otherwise. Fundamentally, this argument concerns the relationship between God's moral perfection and human free will. First, note that God's moral perfection, His holiness, requires His participation in the moral perfecting of fallen human beings. In particular, those who will be united with God for all eternity must become morally perfect (otherwise God would be either unconcerned about moral imperfection 68

10 or unmotivated to help correct the defects, and either supposition is unbefitting an adequate conception of God). This fact is consonant with the Biblical view that the effects of the Fall are to be done away with through the redemptive work of Jesus Christ. It is not just that, according to Christianity, the life of Jesus makes it possible to go to heaven rather than hell, but that through the work of Jesus, the redeemed shall be changed "from glory to glory," coming to participate fully in the nature the One who redeems. So if necessary universalism is true, it is simply impossible that anyone fail to be conformed to the image of God's Son, given that such a failure is incompatible with God's holiness. Yet, such conformance, it would seem, cannot be imposed on us; we must either undertake the task ourselves, or at the very least, acquiesce to the Divine solicitation to aid in the formation of our character. Still, if our cooperation is important, it will also be possible that a person freely choose not to cooperate and forever maintain this uncooperative stance. What possible grounds could be given for thinking that eternal rebellion is impossible? This argument is, I believe, telling against necessary universalism, at least when the type of freedom in question is assumed to be incompatible with determinism. Regarding this assumption, we may note the following. If a defense against McTaggart s Dilemma only succeeds on the assumption that libertarian accounts of freedom are mistaken, we do not have a full response to the dilemma. Hence, even if a defense of necessary universalism were available by employing some other account of freedom, such a defense would be less than satisfactory here. In what follows, then, I will assume a libertarian conception of freedom. I will also assume that some formulation of the Principle of Alternative Possibilities (PAP) correctly describes this notion of freedom (the principle claims that in order to act freely one must be able to do otherwise). There are challenges to PAP stemming from Frankfurt s 14 seminal paper on the subject, but the challenges threaten particular formulations of the principle only, leaving open the possibility of reformulating it to avoid the challenges. For 69

11 example, the original challenge involving a meticulous individual with the power to coerce you to do A, but merely observing your doing A of your own free will, attacks only the version that interprets the phrase can do otherwise in terms of refraining from the action in question. The intuitive response to such counterexamples is to insist that the ability to do otherwise in question must involve something internal rather than external, something like the intention in question. There are complications here that could distract us for a considerable time, but it is sufficient to note what was pointed out earlier regarding the libertarian assumption itself. If we were to adopt an account of freedom that denies PAP, we would have at best a partial solution to McTaggart s Dilemma, since it is far from clear that there is no defensible version of PAP. So here we will assume not only a libertarian account of freedom, but will assume as well that such an account involves some formulation of the principle that in order to act freely one must have been able to do otherwise. This assumption of PAP shows why necessary universalism cannot be salvaged by insisting that there are conditions under which a choice for anything but heaven would be impossible. Consider the following version of such an argument by Thomas Talbott: The picture I get is something like this. Though a sinner, Belial, has learned, perhaps through bitter experience, that evil is always destructive, always contrary to his own interest as well as to the interest of others; and though he sees clearly that God is the ultimate source of all happiness and that disobedience can produce only greater and greater misery in his own life as well as in the life of others, Belial freely chooses eternal misery (or perhaps eternal oblivion) for himself nonetheless. The question that immediately arises here is: What could possible qualify as a motive for such a choice? As long as any ignorance, or deception, or bondage to desire remains, it is open to God to transform a sinner without interfering with human freedom; but once all ignorance and 70

12 deception and bondage to desire is removed, so that a person is truly "free" to choose, 5 there can no longer by any motive for choosing eternal misery for oneself. 1 Talbott maintains that God can transform a person so that they will no longer have any motive for choosing otherwise than in the way necessary for presence in heaven. There are two difficulties with this argument, one having to do with the cognitive transformation posited and the other concerning the relationship between motives and choices. Talbott assumes that God can correct cognitive errors in Belial, getting him to the point of seeing the truth of certain claims that the wicked often ignore or prefer to avoid. It is not clear, however, that God can do so consistent with assumptions about human freedom. Though it is clear that doxastic voluntarism is false, the view that we can simply choose what to believe in any given case, it is far from clear that doxastic involuntarism is true. On this view, the operations of cognition and the operations of the will are never intertwined, so it would never be an intrusion on the operations of the will for God to correct Belial s, or anyone else s, cognitive mistakes. Such a view, however, is not clearly true. There is, at least in certain cases, a sense of optionality regarding what opinion to form or cognitive attitude to adopt. For one thing, evidence is sometimes only vaguely good evidence for a certain claim, giving rise to a feeling of optionality about whether to believe. Moreover, any empirical testing situation triggers the Quine/Duhem thesis that there are always options regarding what to make of recalcitrant experience, and we often have a feeling of optionality regarding what doxastic response to make in such cases. Moreover, the feeling of optionality is even more present concerning large philosophical issues, including issues about what overall perspective on reality to hold. For example, suppose one is a convinced physicalist, denying the existence of God. Must this view change upon experiencing the afterlife? Nothing about such an experience compels any such alteration. For one thing, the affective side of such an individual, including the operation of the will, might overwhelm the operations of the intellect to prevent any such major revision in the 71

13 cognitive sphere, but even on the assumption of a perfectly rational response to such recalcitrant experience, enough changes elsewhere in the noetic structure could accommodate the experience without abandoning physicalism and the accompanying atheism. If such optionality has even the slightest tinge of the will involved, then the assumption that God can correct cognitive errors without threatening human freedom is mistaken and Talbott s argument fails. While I have no definitive argument that the will is implicated by the optionality in question or by the sense of optionality present in some doxastic instances, I also know of no good argument to the conclusion that doxastic involuntarism is true. In the absence of such an argument, the assumption of the argument in the quoted passage above that divine intrusion into a cognitive perspective is unproblematic is uncertain enough to block the force of the argument. There is a worse problem, however, a problem concerning the relationship between motives and choices. Talbott s argument can work only if the absence of a motive for resistance to God is logically sufficient for whatever is necessary for presence in heaven, for without such a claim, the focus on motives in the quoted passage above would be irrelevant to the conclusion that universalism is necessarily true. But if the absence of such a motive is sufficient in this way, then nothing involving freedom can be necessary for presence in heaven (since the presence of such a motive would be incompatible with PAP). Yet, Talbott doesn t deny the need for freedom here. Instead, he intends to argue that we can t make sense of a free choice against God in certain circumstances. Such an argument simply cannot succeed without denying PAP. The argument will have to maintain that if there is no motive and hence no sense to be made of a choice, then that choice is impossible. But such a result undermines even the most internal versions of PAP, and those versions are the only ones plausibly maintained in the face of Frankfurtian counterexamples, so Talbott s argument is successful only to the extent that the kind of freedom in question is not libertarian freedom understood in terms of some internal version of PAP. Hence, it cannot be viewed as part of an adequate response to McTaggart s 72

14 Dilemma. As noted already, however, we cannot give a full response to the dilemma in question without assuming that PAP is a governing principle regarding concerning the type of freedom in question. Given these assumptions, the argument against necessary universalism goes like this. Presence in heaven involves the free submission of the will to that of one s creator, but in any display of freedom, there is the capacity to choose, or try to choose, otherwise. Hence, it is impossible that it be necessary that one freely submit in the way required. Hence it is not possible that it be necessary that everyone ends up in heaven. This argument is valid, and it appeals only to two claims, one involving PAP and the other tying presence in heaven to a something freely chosen. We have already seen why there is no future in the present context in denying PAP, but there are two other ways to consider to try to avoid this argument. There is the premise that ties presence in heaven to free responses, and there is also an assumption about the nature of necessity involved in getting to the conclusion that necessary universalism is false. The latter issue concerning the nature of necessity involved in the free will argument against necessary universalism can be handled quickly. Recall that the problem for contingent universalism is the modal masking of McTaggart s Dilemma, and given this problem, the only hope for addressing that dilemma is to appeal to a notion of necessity that covers all of modal space. Such a notion is plausibly held to be governed by S5, the notion Plantinga characterizes 16 as broadly logical necessity. The particular implication of note in the present context is that such necessity does not change across time: anything possible and anything necessary, in this sense of the terms, is so always and everywhere. So, even if there is some other notion of possibility on which one can lose the power to resist the Divine solicitation, there must be this weaker notion on which such a possibility of resistance remains, on pain of denying PAP. Yet, if 73

15 the possibility of resistance remains, then necessary universalism, in this sense of necessity, is false. That, in a nutshell, is the problem of free will for necessary universalism. Necessary universalism, if true, would solve McTaggart s Dilemma, but in the sense of necessity just outlined, it cannot provide a successful response to this dilemma while at the same time accepting a libertarian account of freedom governed by PAP (clarified using this broadly logical notion of necessity that is also used to characterize the notion of necessity involved in claiming that the salvation of all is necessary) and an account of presence in heaven that makes free choices play an essential role in it. Since we have already argued that the assumption of libertarian freedom governed by PAP is essential to a full response to the dilemma, the only path left open for the necessary universalist is the path of denying that presence in heaven requires a role for free choice. In the next section, I will consider this position, the position that questions the central assumption of freedom in accounting for presence in heaven. For reasons that will become clear later, I term this position ersatz universalism. I will argue that this position offers little hope in avoiding McTaggart s Dilemma, and thus will turn in the following section to the question of whether there is an intermediate universalist position between contingent and necessary universalism that might fare better. First, though, we turn to ersatz universalism. 3. Ersatz Universalism The ersatz universalist grants that necessary universalism cannot answer the free will argument, but insists that necessary universalism can be defended nonetheless. The defense proceeds by first granting the value of freedom, but denying that it is so valuable that it can sustain the possibility of a person ending up in hell. The idea, then, is that God honors freedom 74

16 and solicits our cooperation in securing our presence in heaven, but if such an approach fails, there is always the option of coerced presence in heaven. Consider Talbott s explanation of a view of this sort:...[e]verlasting separation is the kind of evil that a loving God would prevent even if it meant interfering with human freedom in certain ways. Consider the two kinds of conditions under which we human beings feel justified in interfering with the freedom of others. We feel justified, first of all, in preventing one person from doing irreparable harm, or what may appear to us as irreparable harm, to another... We also feel justified in preventing others from doing irreparable harm to themselves; a loving father may... physically overpower his daughter in an effort to prevent her from committing suicide... So... a loving God... could never permit one person to destroy the very possibility of future happiness in another; and... he could never permit his loved ones to destroy the 7 very possibility of future happiness in themselves. 1 According to Talbott, we are allowed free rein regarding our eternal destiny only up to a certain point. That point is the point at which further exercise of our freedom will result in irreparable harm to ourselves or to others. The relevant option in the present context is that of doing irreparable harm to ourselves. The ersatz universalist claims, with Talbott, that there are situations where the impressive value of freedom can legitimately be overridden by other factors, and that one s afterlife destiny is a situation of precisely this sort. If one freely chooses heaven, all is well; but if one does not so choose, at some point the right thing to do, and the thing that God in fact does, is to override our freedom and secure our presence in heaven. Freedom is important, the ersatz universalist claims, but it isn t that important. 18 Before discussing ersatz view directly, I want to distinguish that view itself from Talbott s argument for it relying on the notion of irreparable harm. To tie this notion to the afterlife options of heaven and hell, some assumptions are needed about the nature of hell. For 75

17 consignment to hell to involve irreparable harm, it must be a very bad thing and an inescapably bad thing. Thus, in order for this argument for a limitation on freedom to be accepted, one must already have adopted other elements of the strong view of hell according to which escape from hell is impossible. If we were to suppose, with second chance views, that escape from hell is possible, then there would never be a time at which the envisioned intrusion into human freedom would be warranted, for whatever harm would be involved in the experience of being damned, that harm would be repairable by the change of heart and mind the experience of hell was intended to produce. The logical core of ersatz universalism, however, should not be understood to include any such assumption. If we are to reject the ersatz view, as I will argue we should, we should not reject it simply by claiming that though escape from hell is metaphysically possible, it isn t possible in some weaker sense of possibility (thereby allowing us to claim that the harm in question is never truly irreparable and thus that the point at which a restriction on freedom would be warranted is never in fact reached). So instead of focusing on the argument for ersatz universalism employing the notion of irreparable harm, I will focus on the view itself. Ersatz universalists may characterize presence in heaven in terms of acceptance of God s offer to join him in heaven. There are other options as well, such as conceiving of heaven as a geographical location and imagining salvation in terms of some Divine relocation package. Another alternative is to imagine presence in heaven as involving two kinds of persons: those who want to be there and are happy about it, having so chosen; and those who are not and have not so chosen. But these other options make the view much less attractive, so I ll focus here on the notion of acceptance: some are in heaven as a result of freely acceptance God s offer of salvation while others will be in heaven as a result of a coerced acceptance of God s offer of salvation. Perhaps some will object that my use of the language of coercion is rhetorically prejudicial, but I don t see how it could be. The idea is that God directly and immediately 76

18 secures one s acceptance and thus secures one s presence in heaven, and such immediate and direct securing would clearly involve the use of power to guarantee complaince. But nothing in the discussion to follow will turn on the issue of whether one wishes to include or avoid the use of the term coercion and its cognates. The fundamental difficulty for this account is that it misconstrues the nature of presence in heaven. Traditional Christian doctrine uses the language of heaven and hell to distinguish between those who will love God and enjoy him forever and those who will not. Now, perhaps enjoyment can be passive and imposed from the outside, but love cannot. To love another person requires an expression of agency, and such expressions of agency implicate the notion of freedom. The offer of salvation is, at bottom, the offer of intimacy with the Divine, involving reciprocal expressions of love. To imagine that the expressions in question can be controlled completely from one side of the relationship is to misconceive the nature of love. Love cannot be coerced, but must be freely given. The Biblical imagery of a lover wooing a beloved makes sense in this context, whereas the idea of compelled association is more appropriate to the relationship of master and slave, or even better, that of designer and machine. Given this point, the account we ought to accept about the nature of presence in heaven is one that involves the free union of self with the Divine. Anything less will not be involve the beatific vision itself, but something inferior to it. When we combine this result with the logical notice that there are only two afterlife possibilities, heaven and hell, we are left with the conclusion that forced results of any sort still leave one in hell. It is for this reason that I refer to this position as ersatz universalism. There is a long history in Christian theology of attempting to mitigate the severity of the strong view of hell by making the experience of hell somehow less severe than some of the imaginative portrayals of hell make it out to be. Thus, some adopt second chance views for this purpose and others opt for annihilationism, somehow thinking cessation of existence would be better than the misery of 77

19 eternal conscious separation from God. In other cases, the pain of hell is claimed to be only the pain of loss rather than the pain of sense, so that the suffering of the damned is less troublesome than it otherwise would have been. Whether these views succeed in being mitigations of the severity of the strong view is not the issue here. The issue is, instead, that they are developed for such a purpose, and once viewed in this way, we can see that ersatz universalism is the same sort of view. Instead of hell involving conscious eternal suffering of the worst sort imaginable, the ersatz universalism has God intervening into the will of the unredeemed, changing them in such a way that they can come as close as possible to the experience of heaven, even though they can never quite get there. Even without the experience of heaven, however, they can experience unimaginable happiness, joy, and bliss in spite of the fact that what they are experiencing is not heaven but its logical contrast. They are in hell, but it s just not that bad; they are not in heaven, because the greatest good promised to the redeemed the beatific vision itself is unavailable to them. For those more familiar and comfortable with the position of theological determinism, this argument will hold little persuasive power. For theological determinists hold that presence in heaven is a matter of divine election and decree, completely independent of any libertarian activity of the human will. If one is attracted to such a position, then the account of heaven presented above, involving a turning of heart and mind and will toward the source of all good and truth and beauty, will be perhaps one possible way of loving God and enjoying him forever, but not the only way. Recall that here, however, we are assuming a libertarian account of human freedom in order to provide a complete response to McTaggart s Dilemma. So, whatever the attractions of theological determinism, it is irrelevant here. There is another way to ersatz universalism that I must caution against here. The debate about free will has led to the regrettable practice of referring to various positions in terms of philosophical terminology, so that some talk about libertarian free will and compatibilist free 78

20 will, etc. This language suggests the possibility of their being two kinds of free will, and once one makes this mistake, one can then turn to another, asking which of the two is the more valuable kind of free will. There are no different kinds of free will, however. There is just the ability to do otherwise, which is what human freedom consists in. The libertarian has a certain theory about what the ability to otherwise involves, and the compatibilist has a different theory. The existence of a theory doesn t imply the existence of a kind. In this regard, the following remarks by Peter van Inwagen are worth taking to heart: Having said this about the word able I want to make what seems to me to be an important point, a point that is, in fact, of central importance if one wishes to think clearly about the freedom of the will: compatibilists and incompatibilists mean the same thing by able. And what do both compatibilists and incompatibilists mean by able? Just this: what it means in English, what the word means. And, therefore, free will, incompatibilist free will, compatibilist free will and libertarian free will are four names for one and the same thing. If this thing is a property, they are four names for the property is on some occasions able to do otherwise. If this thing is a power or ability, they are four names for the power or ability to do otherwise than what one in fact does. 19 Van Inwagen makes his point initially in terms of the meanings of the terms in question, whereas I have made the point in metaphysical terms. But both points are in order. One shouldn t let the language of libertarian free will and compatibilist free will confuse one into thinking that there are two different properties or powers or abilities in question so that we could inquire as to which kind of expression of free will is most valuable. There is only the ability to do otherwise, and the argument presented here claims that the union with God involved in the experience of the beatific vision is one that involves the acquiescence to the divine invitation with one s whole mind and will, for only in this way can one count as loving God with one s whole heart, soul, 79

21 mind, and strength. If compatibilism were true, one could do so freely even though determined to do so. We are assuming, in the context of McTaggart s Dilemma, that libertarianism is true, so there is no possibility here about two different expressions of freedom, one having more value than the other. Nor is there any comfort for ersatz universalism in positing different levels of heavenly bliss among the redeemed, with those who love God in the way described above as in places of greater reward than those who have been coerced into heaven. While it may be true that the experience of heaven is better for some than others, there is a non-negotiable requirement here that draws a qualitative distinction between heaven and its contrast. That requirement is the requirement of accepting the divine solicitation and loving him completely. All else is contrast only. Thus ersatz universalism is not really a version of universalism at all, just as fake diamonds are not diamonds and decoy ducks are not ducks. It offers no hope of yielding an adequate response to the free will argument against necessary universalism. In light of this failure, the best strategy for a universalist is to admit the obvious: contingent universalism is too weak to solve McTaggart s Dilemma and necessary universalism is too strong. If there is any hope for solving this dilemma by appeal to universalism, what will be needed is some intermediate position between contingent and necessary universalism. I turn in the next section to the question of whether there is such a middle position. 4. Can A Middle Ground Be Found? If one wishes to find a position stronger than merely contingent universalism, the first natural step to take is to look at counterfactuals for help. Instead of saying merely that none will be lost, a stronger claim turns subjunctive, maintaining that none would be lost. Such a view 80

22 invites perplexity, however, since such a claim involves some implicit condition, and it is not clear what the condition might be. If the full claim is that none would be lost no matter what, we get a view that implies necessary universalism (since the phrase no matter what expresses a universal quantifier reaching every nook and cranny of modal space). So the question is what the implicit condition might be that yields a middle position. One idea is to appeal to is something involving God s creative purposes. If the purpose of creation is for fellowship, and if God is sovereign over his creation, exercising providential control in all of its affairs, it begins to look unsurprising if he would arrange things so that none would be lost. Of course, it can t be impossible for some to be lost, but that is not the claim being made. The claim, instead, is that none would be lost, if creation itself is the product of a desire for fellowship by a sovereign being exercising full providential control. Possibilities of damnation exist, but perhaps in as a remote a sense as the possibility that we are being deceived by a powerful demon so that most all of our beliefs are false. Let us call this position counterfactual universalism. Given this view, we are entitled to claim not only that all will be saved, but that all would be saved if God is loving, sovereign, and exerts providential control over the destinies of humanity. The problem is that such a strengthening isn t a strengthening enough. The same problem that plagues contingent universalism also undermines counterfactual universalism. The problem, it may be recalled, is that contingent universalism only modally masks the problem of hell that underlies McTaggart s Dilemma. The problem, at bottom, is that, whatever the details regarding hell, the design plan is God s and so we can always ask about the design plan and whether it would be wicked to have left (or put) such an option in place. Thus, we may ask, would it be wicked, as McTaggart claims, to consign something to hell? If the worry is that on the strong view, it would be wicked to do so, there is no philosophical consolation in pointing out that no one will in fact be consigned there. But, for the same reason, there is no 81

23 philosophical consolation in noting that no one would be consigned there, so long as God is conceived to be sovereign and exerting control over the course of human affairs. We still have modal masking of the problem occurring; it is just that, to use the language of possible worlds, the range of worlds in which the masking occurs is more extensive for counterfactual universalism than for contingent universalism. The same points hold when one tries to use probabilistic notions and whatever modal implications they may have to sustain a universalist response to McTaggart s Dilemma. For example, suppose one claim that there is no chance that anyone will end up in hell. It is not clear what modal implications there are for such a view, and there are multiple interpretations of the probabilistic notion in question. So, in a variety of ways, such a position is in serious need of clarification. Regardless of how such clarifications proceed, however, we will still be left with the following problem. Standard probability theory leaves open the possibility that things happen that have no chance of happening, and so long as this option remains, the problem of hell underlying McTaggart s Dilemma remains. One might opt for some non-standard approach to probability to close off this possibility, but if one does, then this no chance universalism will imply necessary universalism, and thereby fall to the free will problem for that view. So either way, no chance universalism cannot avoid McTaggart s Dilemma. To avoid the problem of modally masking the problem that underlies McTaggart s Dilemma, we will therefore have to look for a kind of necessity that we can attach to the universalist doctrine in order to be able to claim, in some legitimate way, that modal masking is not occurring anymore. The difficulty in following this path is that the typical notions of necessity that philosophers speak about that are weaker than that of logical or metaphysical necessity won t help at all. For example, in descending order of strength, we might talk about nomological necessity, or counterfactual necessity, or even weaker notions such as psychological necessity or 82

24 the necessity that attaches to some actions in virtue of being in character rather than out of character for a person. None of these options rule out the possibility of creatures for whom damnation occurs in a way that would trigger McTaggart s Dilemma all over again, and thus even if all actual creatures can t be damned consistent with their character or psychology or even the laws of nature, we would yet have no adequate response to McTaggart s claim that such damnation would be wicked. Such notions of necessity thus have no resources to block McTaggart s Dilemma, even if they have the capacity to provide reassurance that no one is under any real risk of being damned. Those familiar with the literature on modal metaphysics, especially in the area of philosophy of religion, will immediately think of another alternative here, however, and it is the notion of necessity introduced by Alvin Plantinga in the context of the problem of evil. Since the problem that underlies McTaggart s Dilemma is a special case of the problem of evil to my mind, by far the hardest version of this problem it would be natural to expect Plantinga s notion of necessity to provide some hope for universalism in its attempt to avoid this dilemma. Though Plantinga did not use this term in describing the view in question, the standard terminology is now to distinguish the class of logically or metaphysically possible worlds from the class of feasible worlds. The class of feasible worlds is composed of those worlds that it is logically or metaphysically possible for God to actualize, and Plantinga s famous argument claims that not every logically or metaphysically possible world is a world that God can actualize. 20 I will call the position that claims that there is no world actualizable by God in which anyone ends up in hell feasible universalism. The claim of the feasible universalist is that, if it is impossible for God to actualize any world in which someone is damned, then we get all the virtues of necessary universalism in replying to McTaggart s Dilemma without the burden of having to deny a libertarian account of human freedom. For the distinction between a possible 83